Anda di halaman 1dari 11

International Journal o/ Pharmaceutics.

21 (1984) 167- 177 167


IJP 00717

A comparative study of the gastrointestinal transit

of a pellet and tablet formulation

S.S. Davis ‘, J.G. Hardy ‘, M.J. Taylor ‘**, D.R. Whalley 2 and C.G.
Wilson 3
I Pharmacy Department. University oj Nottmghom, University Park, Nottin&wn NG7 2RD; ’ Dept a/
Medical Physics. Medico1 School. University o/ Nottmgham, Nottmngh~ and ’ Dept. of Physiolw ond
Phormacoiogy, Medical School. University of Nottingham, Nottingham IV. K.)

(Received March 14th. 1984)

(Accepted May 2nd. 1984)


The gastrointestinal transit of a pellet and tablet formulation have been evaluated
in a group of 6 subjects using the technique of gamma-scintigraphy. Each formula-
tion was labelled with a different radionuclide and the preparations administered
concurrently. The transits of the formulations were found to be highly dependent on
food intake and there was a good correlation between transit times (gastric empty-
ing, arrival at colon) and the calorific value of the meal taken shortly before dosing.
In some cases the pellet system emptied rapidly from the stomach and showed little
spreading in the small intestines..


Controlled release dosage forms such as pellet systems and matrix tablets are
becoming popular in drug therapy as attempts are being made to extend the
biological half-life of certain drugs. The transit of the dosage form through the
gastrointestinal tract can have an influence on the appearance of the drug in the
systemic circulation (bioavailability). The time the dosage form remains in the

* Present address: School of Pharmacy, Leicester Polytechnic, Leicester. U.K.

Correspondtwre: S.S. Davis, Pharmacy Dept., University of Nottinghan.. Umversity Park. Nottingham
NW 2RD. U.K.

stomach is of particular significance and it is now well established that the stomach
handles solid objects greater than 2 mm in diameter in a different manner to smsller
objects or liquids. In addition, factors such as emotional state and diet are important
determinants of gastric emptying (Kelly, 1981). It is sometimes claimed thar a pellet
system is more appropriate for the controlled release of drugs since the individual
particles will be widely dispersed in the gastrointestinal tract. The advent of
gan~ma-scintigraphic methods now allows such speculation to be examined in a
quantitative way and various studies (Casey et al., 1976; Daly et al., 1982; Davis,
1983) have demonstrated clearly the considerable advantages that scintigraphy has
over radiographic procedures for following gastrointestinal transit.
A number of recent investigations (Hunter et al., 1982; Christensen-et al., 1984)
have shown that pellets can sometimes move as a bolus through the intestines and
that such behaviour is highly dependent on the presence of food in the stomach,
The present study was undertaken to follow the transit of a pellet system and a
non-disintegrating tablet, through the gastrointestinal tract of healthy volunteers
following a light breakfast. The two systems were labelled with different radio-
nuclides so that both could be monitored concurrently.

Materials and Methods

An ion exchange resin (Amberlite IRA 410. density 1.2 g/cm”) was sieved to
provide particles in the size range 0.8-1.1 mm diameter and labelled by mixing 1 g
with 50 MBq technetium-99m using 9yn’Tc-sodium pertechnetate obtained from a
generator (Amersham International). Size 3 hard gelatin capsules were each filled
with approximately 200 particles f140 ‘mg) to give an activity of about 2 MBq of
“‘“‘Tc at the time of administration.
The stabilrty of the binding of the label to the ion exchange resin was checked in
vitro under pH and ionic strength conditions relevant to the gastrointestinal tract
f<‘asty et al., 1976; Curt et al., 1980; Hunter et al.. 1982). A further check was made
in viva by monitoring the thyroid glands of the volunteers, since released pertech-
netate would concentrate at this site. Christensen (1984) has described in detail the
in vitro evaluation of coated pellet systems and tests ‘for the integrity of the labelled
preparations. The ion exchange pellets used in the present work had t.he same
gastrointestinal transit behaviour (gastric emptying and intestinal transit} as labelled
~~tt’d pellets (Christensen. 1984) for the same volunteer and for different volunteer
grcjups t Davis, 1983; Christensen et al., 1984).

(_‘oncavc nhlatcs with a diameter of 8 mm and thickness 4.5 mm were prepared

from iactosc in a manner similar to that described by Daly et al. (1982). Prior to
compression the lactose was mixed with a small quantity (20 mg) of a diethylen-
elrian~inepentaacetic acid) (C.1.S. (I.J.K.), London) radiolabe~~ed with 4 MBq in-
&urn-l 11 per tablet.
The tablets were coated with polymethylmethacrylate to prevent their
tion. The integrity of the tablets was checked in vitro under simulated
relevant to the gastrointestinal tract.
The average weight of the tablets was 300 mg and the density 1.39
measured by mercury pycnometry).

Iv vivo studies
A group of 6 healthy male volunteers aged 19-38. height 1.42
60-85 kg, participated with informed consent. Each was allowed to
a breakfast (for example, cereal, coffee or tea, egg. toast. butter, marmala
one hour before the commencement of the study. Each volunteer sw
pellets and tablet with 100 ml of water.
Simultaneous imaging was undertaken of both radionuclides with the ~b~~t~
standing, using a gamma camera having a 40 cm diameter fteld of view and fitt
with a medium energy (300 keV maximum energy) parallel hole collimator, An
external reference marker comprising an adhesive patch labelled with a small
quantity of 99”Tc-sodium pertechnetate was positioned over the liver to the right of
the stomach. Anterior and posterior images each of 60 s duration were taken at
intervals over a period of 24 h and the data recorded by computer for analy&
Subsequently regions of interest were defined around the images of the stomach and
ascending and transverse colon, and the activity in these regions quantified and
corrected for background activity. When undertaking imaging using two radro-
nuclides. correction needs to be made for ‘scatter down’ of the activity of the higher
energy radiation from indium-111 into the energy window of the technetium-99m
photopeak. A correction factor was obtained by administering the indium labelled
tablet and imaging on both channels with indium-111 alone immediately before
giving the technetium-labelled preparation. The counts were then corrected for
radioactive decay. The attenuation of the radiation by overlying tissues can give rise
to incorrect estimates of the amounts of radioactivity in vivo when unidirectional
images are taken (Tothill et al., 1978). The use of the geometric mean of the anterior
and posterior counts gives a result that is approximately independent of the depth of
the source. This procedure was adopted in the present study. The volunteers were
allowed to drink and eat normally during the study.

Results and Discussion

The transit of the pellets and tablet system through the gastrointestinal tracts of
the volunteer group was followed using the gamma camera method. A typical sat of
radionuclide images for one volunteer is shown in Fig. 1. The mdium and tsh-
netium images recorded simultaneously have been combined to show the pellet
system and the tablet in the same view. The tablet and the pellets in the stomach and
the transit through the intestines can be seen. In two views the tablet is seen to be
contained within the pellet mass. The pellets spread to some extent but do not
distribute particularly widely within the intestines. At longer times the tablet is

further advanced in the colon than the pellets. The data have been quantified to
allow the calculation of profiles for gastric emptying and arrival at the ileo-caecal
junction (Table 1 and Fig. 2).
The transit of the pellets in these two different regions of interest is expressed as
the time fq.>r50% to leave or arrive at the particular site. It will be noted that for
some volunteers the emptying of the pellets as well as the tablet formulation from
the stomach was rapid whereas for other volunteers a slower emptying was observed
(Fig. 3). A value for spreading can be estimated from the reciprocal of the gradient
of the gastric emptying or colonic arrival curve (if necessary drawn as a tangent to
the curve at 50% activity) expressed as minutes.
The data for the pellet system are in good agreement with previous investigations
that have employed gamma-scintigraphy to follow gastrointestinal transit (Hunter et
al., 1982; Sjiigren and Bogentoft, 1982; Christensen et al., 1984). The recorded times
for 50% to leave the stomach were between 80 and 180 min for either fasted
volunteers or following a light breakfast.
Data for the emptying behaviour of the individual tablets and their arrival in the
colon are also given in Table 1 (as approximate times since the imaging was not


7h 10h
Fig. 1. Radioisotope images showing the pellet and tablet formulations in different regions of the
gastrointestinal tract. The position of the tablet has been identified by an arrow.

_ . - ..__.-....._._- .-...-- __..

. .. _... -. .-- .- ----- --.-- ---

continuous) and in Fig. 2 as the number of units in a given region at any given time.
In this case emptying is an ‘all-or-nothing’ process that cannot be compared directly
with r.he statistical process for the emptying of a multiple pellet system. However, it
is apparent that the individuals who exhibited slow emptying of pellets also gave a
delayed emptying for the tablet. This interrelationship is sho\h;n in Fig. 4 for both
gastric emptying and colonic arrival. If data for volunteer 1 are omitted (since no
time for .arrival at the ileocaecal junction was obtained) the:n the relationship

1000 6

Time (min)

Fig. 2. Gastrointestinal transit of pellet and tablet systems. A. data from the study of Dew et al. (1982).
Pellet data mean f S.E.M. (n = 6).

0 100 200
I 300
I 100
1 5c0
I 600
I $0
Tim* (mm1
Fig. 3. Gastrointestinal transit of pellets in two different volunteers.
between the data can be expressed as:

correlation coefficient = O.US,n = 10, P < 0.001.

The gastric emptying of single unit dosage form%of similar diame&r to
the present study has bee& examined by various authors
1982; Rosswick et al., 1%7; ~u~~~Li~n~ and Mum, 1
from 30 min to more than 24 h. Recent studies performed by cx
reported gastric emptyin times ranging from OS ta 4.5 h (median 2 h
state and from 2 to 4.5 h (medium 4 h) after a meal (S@gren and
The variation in gastric emptying between indi~d~l$ in the
attributed to idi~~~ratic effects as well as possibk: effects due
natures of the ‘brealifast’ taken before the study. This wilt be di
below. The data on the transit of the pellets (50 arrival at caecum)
through the small intestine indicate a mean trans time from mouth to
about 4-6 h and for stomach to caecum of 3-4 h, Chests et ak
mean values of 246&28 tin (n= 5) and 2042 31 min (n= 8)
through the small intestine of solution and pellet systems. respective&
Table 1 for pellets (227 It:82 min, n J 5) and tablet (188 -f. 23 min. n = 5) are
agreement and indicate that transit in the small intestine is tit&z affected by
or shape of the particles. The few data available for the transit of sin&z uni
forms from stomach (or mouth) to colon are in accord \Icjth the p
(Rosswick et al., 1967; Bechgaard and Christensen, 1982; Wifsm r3t al., I
Dew et al. (1982) have recently studied the transit of tablets felting
(type unspecified) using a radiographic method. Each of 6 volunteers recci


tablets and the recorded data were pooled for purposes of discussion. After 5 h none
of the tablets were found in the stomach, 50% were in the small intestine and 50% in
the colon. After 8 h, 811% of the tablets had reached the colon (‘100% by 24 h). The
two earlier data points have been included in Fig. 2; the mean value (t,,%) of 3?0
min for mouth-to-colon transit for the tablets agrees well with our data in Table 1.
The data obtained in the present work can be related to the known gastrointesti-
nal physiology. The gastric emptying of pharmaceutical dosage forms will depend
upon two major factors; the presence of food in the stomach .and the size of the
dosage form or for a multiparticulate system the size of the constituent units. The
stomach has a remarkable ability to empty different materials in a selective way even
though these materials may be ingested simultaneously (Kelly, 1981). Liquids are
emptied relatively rapidly but digestible materials need to be broken down into
particles of less than 1 mm in size to become suspended in the chyle and then
allowed to pass with the liquids through the pylorus into the duodenum (Meyer et
al., 1979). This in essence represents a sieving of the solid food. Indigestible solids
(such as tablets) of a size greater than about 2 mm are nearly completely retained in
the stomach throughout the postprandial period (Hinder and Kelly, 1977). Only
when the stomach is empty of digestible matter can the intact tablet be emptied by a
special mechanism called the interdigestive myoelectric complex (IMC) (Code and
Marlett, 1975).
Powerful contractions have a role in emptying the stomach of its fasting content
(swallowed saliva, basal secretion of mucus and cellular debris as well as undissolved
tablets). Both liquid and solid gastric content are propelled past the pylorus into the
duodenum. The ability of these contractive waves of the IMC to clear the stomach of
its contents has been called the ‘interdigestive housekeeper’ of the gastrointestinal
tract (Szurszewski, 1969). Consequently an oral formulation administered to a fasted
stomach (or one that has received a light breakfast of low calorific value) may be
emptied rapidly from the stomach. Evidence for this has been presented by Hunter
et al. (1982) and Christensen et al. (1984) for intact capsules and. pellet systems.
After food the stomach has a different emptying pattern that depends on the
quantity and quality of food ingested. Carbohydrates empty more quickly than fats
but when the total energy is considered isocaloric amounts of fat, protein and
carbohydrate empty at similar rates. Thus when meals of different energy content
are ingested, the rate of gastric emptying is such that the number of calories
delivered to the duodenum tends to be constant with time (Chaddock et al., 1974;
Hunt and Stubbs, 1975; Kelly, 1981). In the present study, the good correlation
between gastric emptying of pellets and tablets and food intake can be examined by
estimating the calorific value of the different breakfasts ingested by the’volunteers
(Table 2).
The pattern of transit through the intestines is complex and differs between fasted
and non-fasted states. Material may lie motionless for some time and then be
sequestered into small pieces and pushed to and fro over a short distance within the
intestines. As a consequence, the process of transfer 1s not smooth or uniform
(Weisbrodt. 1981). Hard data on the mixing of intestinal contents are lacking but it
is believed that there is not appreciable exchange of material at the different levels of

the intestines (Gustavsson, 1978). In fasted individuals it is possible to deliver

(radiolabelled) solutions to the ileocaecal junction in an average time of 73 + 7 mi;~
with a range 47 to 140 min (n = 14) (Caride et al., 1982). The IMC again has a role
to p1ay and an *intestinal housekeeper’ can move the intestinal contents rapidly
down the small intestine into the colon (Szurszewski, 1969). A change in intestinal
motility is brought about by eating and hormones released by the presence of
different foods have been impli~t~ (Weisbrodt, 1981). The IMC is either disrupted
or altered.
Read and others (1982) have failed to show any correlation between gastric
emptying and transit time through the small intestine for a large number of normal
subjects after they had eaten the same test meal. The rate of gastric emptying
influenced the transit of food for the first 70 cm of the small intestine only. The
authors suggested that the ileum acted as a buffer region that could delay and
concentrate solid material entering from above. Thus passage through the ileum will
tend to normalize mouth-to-caecum transit (Read et al., 1982). The observations by
Davis (1983). Christensen et al. (1984) that solutions and pellets have similar
intestinal transit times, together with the data in Table 1 showing similar transit
times for pellets and tablets in the small intestine, provide further evidence for the
normalizing role of the small bowel on transit.
The ileocaecal junction forms the boundary between large and small intestines
and is a specialized area of the gastrointestinal tract (Weisbrodt, 1981). Most studies
indicate that the junction maintains closure between the two regions (Cohen et al..
1968) and prevents retrograde passage of material from the caecum (Christensen.
1981). The movement of material in the colon has been reviewed by Christensen


Volunteer Breakfast Cafotific Gastric emptying of pellets Emptying of tablets

VAIC (appiox. time)
150% Spreading value
(kJ1 (min)
(tin) @in)
2 sausages
baked beans,
marmalaoth rade 318% 125 330 600
orange juice 326 30 50 30
fried egg,
tea 2 705 150 105 225
none 0 33 71 42
lea 1046 75 128 30
tea 1950 60 100 60

(1981). The caecum and the sigmoid colon are the major points of delay of material
in mouth to anus transl: (Hansky and Connell, 1962) and in the present studies we
have observed a slow transit of both pellet and tablet formulation in the ascending
colon. During the prolonged residence of intestinal contents in the colon mixing can
occur. For example, Halls (1965) showed, using different batches of radio-opaque
discs of 3 different sizes, that good mixing could occur in the colon even though the
discs were ingested 36, 24 and 12 h before the X-ray examination. Furthermore some
of the latest ingested discs lay in advance of those ingested earlier. The mixing is
believed to occur mainly in the ascending colon (Christensen, 1981).


The gastrointestinal transit of pellet and tablet formulations in healthy volunteers

can be evaluated using the technique of gamma-scintigraphy. Gastric emptying times
and intestinal transit times can be related to the calorific value of the breakfast
ingested before the study. The average transit times for pellets and tablets in the
small intestine are similar and agree well with other data on pellets and solutions
(Davis, 1983). The transit behaviour of the formulation can be understood by
reference to the known physiological properties of the gastrointestinal tract, in
particular the interdigestive myoelectric complex.


The authors wish to thank Dr. M. Wastie. Department of Radiology. Queen’s

Medical Centre, Nottingham, for advice and helpful discussions.


Bechgaard. H. and Christensen, F.N.. Physiological factors and controlled rcleose dos;tge forms. Pharm.
J.. 229 (1982) 373-376.
t’asey, D.L., Beihn, KM.. Digenis, G.A. and Shambhu, M.B.. Method for monitoring hard gelatin capsule
disintegration times in humans using external scintigraphy. J. Pharm. Sci.. 65 (1976) 1412--1413.
Caride. V.J.. Prokop. E.K.. McCallum. R., Troncale. F.J.. Buddoura. W. and Winchenbach. K.. Scinti-
graphic evaluation of gastrointestinal trans_t time. Comparison with hydrogen breath test. Nuclear
Medicine and Biology, pp. 2427-2429. Proceedings of the Third World Congress of Nuclear Medicine
and Biology, Paris, 19X2.
C‘haddock. T.E., Carlson. G.M. and Hamilton. C.L.. Gastric emptying of a nutritionally balanced Itquid
deep in the rhesus monkey. In Proceedings of the Fourth International Sympos~ium on Gastrointestinal
Motility. E.E. Daniel(Ed.), Mitchell Press. ‘lancouver. 1974. pp. 514- 522.
C‘hrtstensen. F.N., In vitro and in vivo evaluation of Iahelled pellets. in prrparatwn.
C’hrlstenxen. F.N.. Davis. S.S.. Hardy, J.G., Taylor, M.J.. Whallry. D.R. and Wilson. C’.ci.. The use (,I
gamma scintigraphy IO follow the gastrointestinal transit trf pharmaceutical fnrmulatinns. J. Pharm.
Pharmacol. submitted.
C’hrlstensen. J.. Motility d)f the Colon. In L.R. Johnson (Ed.). Phystolngy of the Gastrointestinal Tract,
Raven Press. New York, lY81. pp. 445-471.
Code, C.F. and Marlett. J.A., The interdigestive myoelectric complex of the stomach and UII.&J
dogs. J. Physiol. (Land.). 246 (1975) 289-309.
Cohen, S., Harris, L.D. and Levitan. R.. Manometric characteristics of the human rlc.ocecai juncttouaf
zone. Gastroenterolcgy. 54 (1968) 72-75.
Curt. N.E., Hardy. J.G and Wilson, C.G.. The use of the gamma camera to study &e 8atiromtecunaJ
transit of model drug formulations. In Drug Measurement and Drug Effects m
Science, 4th Int. Coil. Prospective Biology, Pont-a-Mousson 1978. Karger. BaseI.
Daly, P.B.. Davis, S.S., Frier. M., Hardy, J-G.. Kennerley. J.W. and Wilson.
assessment of the in vivo dissolution rate of a sustained release tabiet. Int. 1.
Davis, S.S.. The use of scintigraphic methods for the evaluation of drug dosage forms in the
nal tract. In Breimer. D.D. and Speiser. P. (Eds.). Topics in Pbarmaceu ttcal Screnccs 1
Biomedical. Amsterdam, 1983. pp. 205-215.
Dew, M.J., Hughes. P.J.. Lee, M.G.. Evans, B.K. and Rhodes, J.. .An oral preparation to &ease drugs m
the human colon. Br. J. Clin. Pharmacol.. 14 (1982) 405408.
Gustavsson, S.. Studies on the transport of small bowel contents. 1. Med. Sci.. 83 (1978) 167-I
Halls. J.. Bowel content shift during normal defaecation. Proc. Roy. Sot. Med.. 58 (1965) 859-
Hansky. J. and Connell. A.M., Measurement of gastrointestinal transit using radtoactive chromrum. Gut.
3 (1962) 187-188.
Hinder, R.A. and Kelly, K.A.. Canine gastric emptying of solids and hquids. Am. J. Physrok. 233 (1977,
E335- E340.
Hunt. J.N. and Stubbs. D.F.. The volume and energy content of meals ;15 detsmunants of wtrrc
emptying. J. Physiol. (Lond.). 215 (1975) 209-225.
Hunter. E.. Fell. J.T. and Sharma. H.. The gastric emptying of pellets contained in hard gelatrn capsub.
Drug Devel. Ind. Pharm.. 8 (1982) 751-757.
Kelly. K.A.. Motility of the stomach and gastroduodenal junction. In L.R. Johnson (Ed). Physi xf
the Gastrointestinal Tract. Vol. 1. Raven Press. New York. 1981. pp. 393-410.
Meyer, J.H., Thomson, J.B.. Cohen, M-B.. Shadchehr. A. and Mandiola. S.A.. Sieving of sobd food bv the
canine stomach and sieving after gastric surgery. Gastroenterdogy. 76 (1979) 804-813.
Muller-Lissner. S.A. and Blum. A.L.. The effect of specific gravity and eaung on gastric ernpt)mg of slow
release capsules. New Engl. J. Med., 304 (1981) 136: - 1366.
Read, N.W.. Cammack. J.. Edwards. C.. Holgate. A.M.. Cnnn. P.A. and Brown. C.. Is the transit time of a
meal through the small intestine related to the rate at which it leaves the stomach? Gut. 23 tl982)
Rosswick. R.P., Stedeford, R.D. and Brooke. B.N.. New methods of studymg mtestmal transtt tImA. Gut.
8 (1967) 195-197.
Sjogren, J. and Bogentoft. C., How useful is enteric coating to improve drug dellvery? In Buodgaard. H..
Hansen, A.B. and Kofod. H. (Eds.). Opdmisation of Drug Delivery. Copenhagen. Munksgaard. 1982.
pp. 53-62.
Szurszewski. J.H.. A migrating electric complex of the canine small intestine. Atn. J. Pbysioi.. 217 (1969)
Tothill. P.. McLaughlin. G.P. and Heading. R.C.. Tcuhniques and error m scmtigraphic measurements &
gastric emptying. J. Nucl. Med.. 19 (1978) 256-261.
Weisbrodt. N.W.. Motility of the small intestine. In L.R. Johnson (Ed.). Physrokrgv of the Gastromtcstt-
nal Tract. Vol. 1. Raven Press, New York, 1981. pp. 411-443.
Wilson. C.G.. Parr, G.D.. Kennerley. J.W.. Taylor. M.J.. Davis. S.S.. tiard_r. J.G. and Res. J.
Scintigraphic and in vivo pharmacokinetrc monitorin on a sustarnrd r&case acctylsalic\lic a&
formulation. Int. J. Pharm.. 18 (1984) l-8